Talk by Kent Wittenburg (MERL)

On recent projects in information visualization from Mitsubishi Electric Research Laboratories.

Title: Information Visualization Research at MERL: Bars and Seaweed
Dr. Kent Wittenburg, PhD, Mitsubishi Electric Research Laboratory Headquarters (Vice President & Director)

Date: May 24th, 2010
Time: 12:00-13:30
Place: Universidad Carlos III de Madrid. Avda de la Universidad 30. Leganés. Rey Pastor Building (Library Building). Room 3.3.B01.
Local host: Alessio Malizia

The participation is free.

Abstract:

This talk will begin with a short overview of Mitsubishi Electric Research Laboratories followed by highlights from two recent projects in information visualization. The first project involves interactive visualization methods for set-valued attributes for parallel bargrams (equal-height histograms). Parallel bargrams have proved of value for exploratory visualization with many types of relational data in a wide variety of applications, but, so far, set-valued attributes have not been successfully integrated into such systems. We propose a method for doing so. The second project, joint with Cliff Forlines, is an exploratory foray into the use of 3D shapes in representing spatial-temporal data. It is well-known that humans have well-developed cognitive abilities to perceive shapes in 3D regardless of orientation. We are exploring the possibility that recognizable shapes that represent complex patterns in space and time can be constructed from multidimensional sensor data. Our basic 3D shape metaphor is extruded radar graphs that we call Wakame (Japanese for seaweed), as they are evocative of columns of seaweed growing from the ocean floor. Our motivating example is management of energy efficiency in building environments, a case in which the spatial characteristics of the multidimensional sensor data are relatively static but the temporal range may be extensive. We have identified such patterns as a failed HVAC component revealing itself through cascading waves of rising temperatures (bulges) in neighboring spaces (Wakame). We also relate these 3D views to more conventional 2D graphs through animated transitions and incorporate methods for discovering anomalies in the data.

Short Bio:

Kent Wittenburg has had a 25 year career in corporate research including stints at MCC, Bellcore, GTE/Verizon Labs, and Mitsubishi Electric Research Labs. He currently is a VP and Director at MERL and manages research in multimedia, digital communications, and data analytics. Among his current research interests are multivariate information visualization, visual (RSVP) interfaces, and multimodal (speech + touch) interfaces for automotive environments. He holds a PhD in computational linguistics from the University of Texas at Austin.

Contact us

The Digital Living Initiative

Instituto de Cultura y Tecnología
Universidad Carlos III de Madrid

Calle Madrid 126
28903 Getafe, Spain

E-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it.
Phone: +34 91 624 9456
Fax: +34 91 624 9129